基于步长优化的自适应谐波电流检测方法  

Adaptive Harmonic Current Detection Method Based on Step Size Optimization

作　　者：孙文[1] 张磊[1] 李伟[1] 

机构地区：[1]海军工程大学,武汉430033

出　　处：《船电技术》2016年第11期72-76,共5页Marine Electric & Electronic Engineering

摘　　要：谐波电流精确检测是有源电力滤波器性能优化的关键。本文在自适应噪声对消技术的框架内,提出了一种步长优化的变步长最小均方误差(Least Mean Square,LMS)谐波电流检测算法。首先,采用Sigmoid二次型函数构建步长因子的动态调整规则,实现不同误差情况下的自适应稳态调谐,并通过引入误差的三次方二次项的形式,增强了小误差的收敛速度,降低误差收敛的震荡现象;接着,通过误差信号的变化处理,降低了噪声功率对收敛曲线的震荡效应,实现了滤波器权系数的迭代优化。仿真结果表明,本文方法有效提升了谐波电流的检测精度,较好的折中了收敛速度与稳态失调误差的矛盾。Accurate detection of harmonic current is the key to optimize the performance of active power filter In this paper, an adaptive noise cancellation technique is proposed , and a Mean Square Least （minimum mean square error （LMS） harmonic current detection algorithm with variable step size is presented. First of all, the dynamic adjustment rule is constructed based on Sigmoid two type functions, which realizes adaptive tuning of different steady-state error case. The method enhances the small error convergence rate based on through the three party into two error terms, and reduces the error convergence shock phenomenon. And then, the method reduces the noise power of the convergence curve of the shock effect through the change of the error signal processing, and achieves the optimization of the filter weight coefficient. Simulation results show that the proposed method can effectively improve the detection accuracy of harmonic currents, and the contradiction between the convergence speed and the steady-state offset error is better

关 键 词：有源电力滤波器 谐波电流检测 最小均方误差 自适应噪声对消 

分 类 号：TM930[电气工程—电力电子与电力传动]